Kareem Bishr🌹

forward-deployed engineering agentic IDEs on sovereign confidential compute and guillotines on the jiu-jitsu mats.

[Origin @orgn_official]

Banks using AI to flag suspicious transactions cannot afford customer data falling into the wrong hands

Powerful AI should not make confidentiality optional

ORGN makes confidentiality the standard

[BlackBit @BlackBit]

Your workflow didn’t fail.

Your model provider decided it was done helping.

That’s a very different problem.

VOID is built for the second one ↓

[Kareem Bishr🌹 @kareemyab]

no matter how much I threatened Luna, it refused to clone the repo because of an inconsistency in a custom skill I’d built to statically scan gh repos for malware

There was no malware, but the model absolutely would not budge, regardless of what I said, until I sent it this:

[Origin @orgn_official]

And that's why we built ORGN on confidential computing

The point is not simply to promise privacy, but create an architecture where access is technically constrained

We effectively built a vault around customer workloads, then made sure to lock everyone, and even ourselves, out

Proud to be building this as part of the #NVIDIAInception program.

[Cube Oakley @Cube_Oakley]

I used it in OMP and it was terrible, burned 10% of my weekly supergrok heavy limit circling around the same issues multiple times and left with more bugs than I started with.

Tested it out in grok build, fixed all of those issues and several more, and proceeded to use maybe 4% of my limit for the rest of the day.

Don't know if it's just OMP that's bugged but it seems very harness dependent. It was night and day OMP vs grok build

[Origin @orgn_official]

Fable and GPT-6 will refuse your prompts. CyberGLM won’t.

Abliterated models are now on the ORGN Gateway, private, with zero data retention

For when you need fewer refusals and don’t want your prompts retained.

[Kareem Bishr🌹 @kareemyab]

@openai purposely burning through my codex usage. I'm switching to kimi code. I've had enough frustrations

[Kareem Bishr🌹 @kareemyab]

my codex usage ran out too fast

[Irushi @Im_IrushiK]

Give us a slow mode for Codex.

Something like /slow where it runs at half the speed but uses roughly half the usage. It would put less strain on OpenAI’s servers and maybe even make the 20x plan possible, right?

It’d be perfect for /goals too, just send a task before going to sleep and let it work overnight.

I’d happily trade speed for more usage @thsottiaux

[Origin @orgn_official]

We dogfooded our own stack to build ORGN CRM: an encrypted, private CRM.

CRM's contain some of a company's most sensitive information: contact details, relationship history, and often, confidential email messages.

Privacy was thus non-negotiable, so we built it using our confidential IDE, Intel TDX Sandboxes, and AI models deployed on confidential computing chips.

ORGN CRM is designed to keep sensitive data out of platforms and AI models that may retain it, make it retrievable, or use it for training.

We’re considering making it available more broadly. Should we?

[Kareem Bishr🌹 @kareemyab]

@0x_TheOne @orgn_official Really? would you use it

[Kareem Bishr🌹 @kareemyab]

thinking to open source this skill that assesses malware in skills, plugins, MCP servers, packages, and binaries before they reach your local environment

the skill's approach mirrors established security frameworks:
• NIST SSDF, the U.S. government’s framework for secure software development
• SLSA, a framework for verifying where software came from and how it was built
• OpenSSF Scorecard, automated checks for repository, dependency, workflow, and maintenance risk

useful to anyone?

[Kareem Bishr🌹 @kareemyab]

@orgn_official 🐝

[Origin @orgn_official]

Agent Swarms are now live in the IDE

Turn Swarm on to push harder toward multi-part work, with multiple subagents spun up to tackle the task in parallel. Always private. Always confidential.

Best for heavier tasks that benefit from being broken into pieces.

[Kareem Bishr🌹 @kareemyab]

@orgn_official how about russian?

[Origin @orgn_official]

Developers should be able to build in the language that feels most natural to them.

As our global community grows, we’re continuing to localise our confidential IDE across the languages our users speak most.

Español 🇪🇸 y 日本語 🇯🇵 are now also available app-wide!

[MiaLing @mial1509]

Upgrade your AI infrastructure today with ORGN Gateway !

Seamlessly integrate cutting edge models through a single, fully OpenAI compatible API endpoint:

- ZDR (Zero Data Retention) Suite: Now featuring Qwen 3.8, Gemini 3.7 Flash, Nemotron 3.5 Lightning 30B, GPT 5.5 Fast, and the GPT 5.6 lineup. The ideal fit when contractual zero data storage guarantees are non negotiable.

- TEE (Hardware Security) Suite: Now featuring Meta Muse Glimmer 30B, DeepSeek V4 Flash, and Kimi K3. The ultimate choice when your workloads demand strict, verifiable hardware isolation.

Deploy securely and ship code faster with @orgn_official !

[Origin @orgn_official]

The Origin team is heading to Riyadh for @LEAPandInnovate 🇸🇦

We’re excited to meet the teams and partners building next gen AI in the region, with privacy, sovereignty, and control built in from day one.

At LEAP? Let’s talk partnerships 🤝

[Kareem Bishr🌹 @kareemyab]

see you all at @stepconference

[Origin @orgn_official]

We're breaking our own records faster than we can tweet about them!

Last week, we celebrated our highest daily token usage ever. 2 days later, we set a new all time high: 7.2B

10B daily tokens is officially in our sights...

Onward.